Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.
Artificial Neural Networks (ANNs)
Become skilled in Artificial Intelligence, Machine Learning, Generative AI, Deep Learning, Data Science, Natural Language Processing, Reinforcement Learning, and more with this complete 0 to 100 roadmap repository. Designed for beginners and experts alike, this project offers a comprehensive learning path, curated resources, hands-on tutorials, and
A unified hub for the Scratch Series — 200+ algorithms across ML, DL, NLP, LLM, GenAI, and TimeSeries, all implemented from scratch with NumPy.
🧠 Master AI and Machine Learning from scratch with this comprehensive roadmap featuring curated free resources for all skill levels.
A 125M-parameter decoder-only language model built from scratch in PyTorch, featuring a custom BPE tokenizer, Rotary Positional Embeddings (RoPE), Grouped-Query Attention (GQA), SwiGLU activations, and sharded dataset streaming. The project covers the full training pipeline, from large-scale pretraining to supervised fine-tuning (SFT).
A 55M-parameter production-ready Seq2Seq chatbot built from scratch in PyTorch. Features a Bidirectional LSTM Encoder, Bahdanau Attention, LSTM Decoder, and SentencePiece tokenization. Optimized via uv with Automatic Mixed Precision, PyTorch sharded dataset streaming, advanced beam search decoding, and comprehensive checkpoint resumption.
To associate your repository with the ai-from-scratch topic, visit your repo's landing page and select "manage topics."